The world of dog training is fraught with dogma and “religious wars”. I understand some of the differences. Whether you believe violence and aversion-based techniques have a place in training a sentient animal is a basic ethical question. But other conflicts have me more perplexed.
A case in point is an online group about reactive dogs that will only tolerate discussions of desensitization and counter-conditioning (DS/CC). Let me be clear that DS/CC is an essential tool for positive training, especially for reactive dogs. But I don’t think it’s the only useful tool, and I think it has limits that must be complemented by other techniques.
The DS/CC are critical of what they call avoidance techniques. Although I haven’t seen a definition of this term, I think they mean teaching a dog to move away from something that makes it uncomfortable. Their argument usually goes something like this: You’ve taught a dog to avoid children (which scare it) instead of desensitizing the dog to kids. You have some visitors and the dog finds itself in close quarters with a child and can’t get away. It will react, and possibly bite the child.
Fair enough, but depending on DS/CC assumes that a dog has a limited number of triggers that you can densensitize it to. My counter argument would go like this: Your human-reactive dog encounters a person who is different from anyone it’s ever seen (someone on crutches, wearing a big winter parka, shuffling their feet…some dogs are freaked out by seemingly small changes). Since it has never learned to move away from Scary Things, it reacts, possibly biting the befuddled person who doesn’t understand what s/he has done.
This is why I think it’s important combine positive techniques to give the dog the best repertoire of coping mechanisms possible. I think teaching avoidance is especially important for dogs that were poorly socialized as puppies, especially dogs who come from puppy mills, pet stores, etc. My reasoning is based on the concept of learned helplessness: when animals find it impossible to escape from something aversive, they eventually stop trying and just resign themselves to the thing they find aversive. (In the original experiments establishing this principal, the aversion was actually an electrified floor, which the dogs stayed on instead of hopping over a low barrier to safety because they had learned there was no escape in earlier phases of the experiment.)
So imagine you are a puppy mill dog who never gets to leave its crate. When scary people come towards you, there’s nowhere to escape, and you eventually stop trying. At this point, Flight or Fight is reduces to one option: Fight, which becomes your default reaction to every threat.
That’s why I think it is so important to explicitly teach a reactive dog that moving away is possible and can be effective. But it shouldn’t be the only thing you’re teaching him.
I think that DS/CC and avoidance are not only complementary, they’re convergent. DS/CC has reduced our need for avoidance, but it has also reduced the distance for avoiding triggers. It used to be hard for Spencer to walk down the opposite sidewalk from a stranger. Now we can see someone coming down the same sidewalk and just swerve around them. Knowing in advance that he doesn’t have to deal head-on with a stranger (which is a very confrontational approach for a dog), has made it easier to desensitize at ever closer quarters.
I consider it a victory when Spencer doesn’t even pay attention to a nearby Scary Person. (The result of DS/CC.) But I also consider it a victory when we come around a corner and discover in front of us and Spencer automatically moves to the left or right to avoid that person. (The result of avoidance.) In both cases, he’s chosen a more acceptable coping mechanism than lunging and snarling.